The PRIMED Consortium: Reducing disparities in polygenic risk assessment.

Area of Science:

• Genetics
• Public Health
• Bioinformatics

Background:

• Polygenic risk scores (PRSs) offer potential for disease risk prediction and prevention.
• Current PRSs exhibit performance disparities across ancestries due to historical genomic data biases.
• This can exacerbate existing health inequities in clinical applications.

Purpose of the Study:

• To develop and implement methods for improving PRS performance in diverse global populations.
• To address health disparities arising from underperformance of PRSs in underrepresented groups.
• To foster equity in the development and application of polygenic risk assessment.

Main Methods:

• Aggregating and harmonizing diverse phenotype and genotype datasets on the AnVIL cloud platform.
• Utilizing population and statistical genetics approaches for individual- and summary-level analyses.
• Developing a common data model for cohort studies and electronic health records.

Main Results:

• Established methods for secure data sharing on AnVIL.
• Created a harmonized data model for global cohort and EHR data.
• Adapted population descriptor guidelines for international cohorts.

Conclusions:

• The PRIMED Consortium is advancing equitable PRS development through data harmonization and advanced analytical methods.
• Joint modeling of social determinants of health and PRSs is being investigated for comprehensive risk prediction.
• Efforts focus on improving PRS utility for cardiometabolic diseases and cancer across diverse populations.